(adapted from the applicant's abstract): A prominent hypothesis in neuroscience is that synaptic plasticity underlies learning. Vocal learning in song birds offers several advantages for addressing this question: Birdsong is a learned behavior that is acquired and practiced during two relatively discreet developmental periods, and the end product of learning is a stable, stereotyped motor act. To examine potential cellular mechanisms of song learning, physiological and behavioral analyses will be combined in individual birds. Comparing biophysical properties of glutamatergic synaptic transmission in zebra finch nucleus RA with a behavior measure of plasticity, the hypothesis will be tested that a reduced contribution of the NMDA receptor mediated component of synaptic transmission causes crystallization. In addition hormones will be used to manipulate the time course of crystallization by castrating juvenile finches (to prolong variability) or implementing testosterone (to accelerate crystallization). If NMDA receptors are important altering the timing of crystallization should concomitantly alter biophysical changes. Finally, subunit-specific NMDA receptor antagonists will be used to test the hypothesis that a switch in subunits causes the decreased contribution of NMDA receptors to synaptic transmission.